Multilingual Scenario
Multilingual scenarios in natural language processing (NLP) focus on developing models and techniques that effectively handle multiple languages simultaneously, aiming to overcome limitations of monolingual systems and improve accessibility for global users. Current research emphasizes leveraging advanced transformer architectures, contrastive learning, and knowledge distillation to enhance model efficiency, accuracy, and robustness across diverse languages and tasks, including text spotting, speech recognition, and machine translation. This research is crucial for advancing multilingual NLP applications, such as cross-lingual information retrieval, multilingual chatbots, and ethical AI development that mitigates biases inherent in predominantly English-centric datasets.
Papers
HEALTH-PARIKSHA: Assessing RAG Models for Health Chatbots in Real-World Multilingual Settings
Varun Gumma, Anandhita Raghunath, Mohit Jain, Sunayana Sitaram
Better to Ask in English: Evaluation of Large Language Models on English, Low-resource and Cross-Lingual Settings
Krishno Dey, Prerona Tarannum, Md. Arid Hasan, Imran Razzak, Usman Naseem
The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm
Aakanksha, Arash Ahmadian, Beyza Ermis, Seraphina Goldfarb-Tarrant, Julia Kreutzer, Marzieh Fadaee, Sara Hooker
Improving the Consistency in Cross-Lingual Cross-Modal Retrieval with 1-to-K Contrastive Learning
Zhijie Nie, Richong Zhang, Zhangchi Feng, Hailang Huang, Xudong Liu